BACKGROUND CONTEXT: Global balance of human standing is analyzed as the geometric sum of the individual alignments extending from the spinal column to the pelvis, and to the lower limbs. The innovative EOS system has opened new perspectives for the global analysis of whole-body alignment, but its use is very limited because of its high cost. An alternative may be to combine the whole-spine lateral radiograph and the lateral scanogram in the global sagittal analysis of whole-body alignment.
PURPOSE: We examined to determine the validity and reliability of a lateral scanogram in the measurement of sacropelvic parameters.
STUDY DESIGN/SETTING: A retrospective radiological study was carried out.
PATIENT SAMPLE: We randomly selected 100 sets of digital radiographs, both whole-spine radiographs and lower-limb scanograms, from our database.
OUTCOME MEASURES: Sacropelvic parameters, including pelvic incidence, sacral slope, and pelvic tilt, were measured on both whole-spine radiographs and lower-limb scanograms by three independent examiners on three separate occasions.
METHODS: Agreement regarding the measurements on both image types was calculated to assess the validity of the lateral scanogram for use in whole-body alignment determinations. Intraobserver and interobserver reliabilities among the types of measurements were calculated.
RESULTS: The sacropelvic area on the lateral scanogram was not visible in 19 patients (19%). In the remaining 81 patients, the sacropelvic parameters on the lateral scanogram were similar to those on the whole-spine lateral radiograph (Pearson correlation coefficient, 0.764-0.805). Intraobserver and interobserver reproducibilities for both modalities were good to excellent (intraclass correlation coefficient, 0.657-0.984).
CONCLUSIONS: Sacropelvic parameter measurements on lateral scanogram were reliable and were similar to those measured on whole-spine lateral radiograph. Thus, global alignment can be evaluated using the lateral scanogram in combination with the whole-spine lateral radiograph.